Moonlight wrote: ↑
Tue Feb 11, 2020 7:36 pm
It was the opposite, how to create Topographic contours from any mesh. By the way, how many programs that have not implemented this option !!
My bad! i was so excited by the prospect of it that i didnt actually read the document. The video you uploaded is beyond amazing though. Around minute 44 is particularly mind chattering. But Also, it helps to illustrate the following:
-Deconstructivism, as a philosophical and formal method to create the built environment is as death as disco (i might be wrong on this, but i doubt it). Many of us appreciate it for its will to explore the limits of materials and non regular geomtrical forms. But as the resources of the world become more scarce, the odds of seeing this kind of archicteture get built seem less and less likely every year. I ocasionally like to dance some disco on a given Friday night though!
-Its (arguably) spiritual succesor, generative and organic parametric design, seems even more grand in its ambitions while being even less rooted in the economic and urban realities in which it pretends to insert itself.
-Archicad struggles a lot if you want to design deconstructivist stuff and basically cannot natively handle Organic/generative design
-Not because they are heavily promoted in magazines and whatnot, it means generative design is/should be the norm. Quite the opposite in fact. Im confident to estimate the amount of comissions that require organic/generative designs account to much less than 1% of the whole universe of comissions. This is specially true in less devolped countries where it would be much much much less than 1% as other needs require everyone atention. (if anyone has hard data on this, please share).
-The last point doesnt mean that just because we can´t/should/dont know hot to do it now, we shouldnt do it in the near future. Preparation is key to survival
-A self-build social housing project in south America needs different tools and construction processes than a Zaha Hadid one in Dubai (may she rest in peace). Why should it be different for software?
-Some offices seem to need super detailed real world objects for their projects. Other offices are happy with the ability to simbolicaly represent objects as long as they can define their dimensions, position in space, specifications, can be listed and it represents in plan, section, elevation and 3d.
-Super detailed objects tend to increase file size?
So to wrap all this rant up, and returning to the Pareto principle, i think Graphisoft would need not 80% but more than 90% of their effort, resources and manpower to create an in-house parametric modeller. From a business point of view it would make absolutely no sense to spend that much for something that would be used in real world situations by around 1% of its user base (i would say much less than that), and i doubt they would be able to compete with the best of the best in this area which is grasshopper and rhino.
Usage percentage may vary, but the same can be said about a GDL Graphical user interface.
Innovation usually comes from top to bottom. Catter to the rich big companies, the masses will follow.
But sometimes, every once in a while, you have to keep the masses happy.
So here are some cost-effective ideas. Probably they already exist, but who knows:
-For the GDL object creation, asuming a GUI is out of the question for now, Why instead of all of us wanting every manufacturer on earth spend precious resources on modelling their products and updating them every year, which, in the case of GDL, it has already been established by almost everyone that it is no walk in the park.
So instead of that, why not create GDL objects that can absorb the specifications (dimensions, ID, description, resumed and full specification, center, point of conection ETC) from a GLOBAL list of manufactured products which would be much easier to mantain and edit. And by Global i mean a Global Database of manufacturers and specifications from all over the world, accesible from withing Archicad.
Example: A bed is a bed everywhere i look, but dimensions vary. What if i want an specific Crate and Barrel bed? I put my global Bed object, and then, within the object, i pull down my Manufacturer list, all of it perfectly categorized, listed and daily updated via online connection to a global database maintained and curated by a reputable institution, select Create and Barrel, the model i want and then PRESTO!, the global object changes itself to the specification of that particular model, WxDxH measures, specifications, ID, etc. Yes, it will not be the EXACT real world model, but you will have its exact dimensions and specifications and they will update automatically (if the model ever goes of production, it will alert you on that). Sweetness.
-For the above to be done, a keynote system management tools is imperative.
Yet, people might still need custom based GDL objects. Mr. Moonlight mentioned in one of his posts that if more and more data is added, complexity arises. Mr. Clarence mentioned that teamwork can be pushed far beyond of what it is now.
So, How about a platform for Archicad projects where everyone can share what they want (old projects, pln files, resources etc, even an online collaborative environment where many people can work a given project just for fun (as if everyone had time for that). Something like GitHub but for archicad users. But most importantly, put together in one place a very scarce resource: GDL programmers.
Then, all the installed Archicad user base would have access to that, and for a very reasobable price we can download and request custom made GDL objects. This way everyone benefits: Normal users get access to the beauties of GDL programming while GDL programmers get fairly paid for their work. Graphisoft would just need to initially set up and fund the platform and then sit back and relax. Economics of scale at work and verybody wins, hopefully.